Rajesh Power Services Ltd 544291

Summary

Rajesh Power Services is an Ahmedabad-based power transmission and distribution contractor. It traces its history to 1971 and listed on the BSE SME platform in December 2024, after an IPO priced at ₹335. Most of its work is underground cabling, medium-voltage covered conductor (MVCC) conversion and substations for Gujarat's distribution and transmission utilities. It holds a Class "AA" electrical contractor licence in Gujarat and Class "A" in Rajasthan and Madhya Pradesh. It also owns 25.48% of HKRP Innovations, a smart-grid software company. In FY26 it signed its first battery storage project: a 65 MW / 130 MWh asset it will own, with a 12-year tariff contract from GUVNL. The founding Panchal and Patel families hold 72.70%.

Growth has been fast. Revenue rose from ₹207 crore in FY23 to ₹1,628 crore in FY26. FY26 alone was up 52%, with EBITDA of ₹197 crore (a 12.1% margin) and consolidated net profit of ₹143 crore. ROCE is about 44–47% depending on the source. The order book plus L1 bids stood at ₹3,326 crore in March 2026 and ₹3,742 crore in June. CRISIL moved its A- rating to a Positive outlook in July. Management is guiding to about 40% revenue growth in FY27 at an 11–12% EBITDA margin.

None of that profit has turned into cash. Standalone operating cash flow was −₹22 crore in FY24, −₹26.7 crore in FY25 and −₹41.4 crore in FY26. In FY26, working capital absorbed ₹212 crore, equal to 38% of the year's revenue increase. Receivables plus retention money reached 141 days of revenue. The gap was filled by stretching suppliers, including ₹86.8 crore owed to MSMEs (up from ₹4.9 crore) and ₹86.5 crore under bank supplier-finance arrangements, and by ₹49 crore of extra borrowing. The share price has fallen from a 52-week high of ₹1,605 to ₹733, and at that price the stock trades on 9.6x trailing earnings.

Our rating is BUY, with a target of ₹945 (+28.9%). Even after assuming growth well below guidance, a heavy working-capital drag, a 14% cost of capital and a deep discount to peers, both our methods value the shares above today's price. The discount is still justified to a large degree. Customer and geographic concentration is high, the related-party arrangements are extensive, and the annual report contradicts itself in several places. If working capital keeps absorbing 38% of incremental revenue, as it did in FY26, our DCF falls below the current price. We call this a high-risk BUY.

• The market is pricing Rajesh Power as if its earnings will never turn into cash. At 9.6x trailing and about 7.4x our FY27 base-case earnings, the stock trades at a 35–63% discount to listed T&D EPC peers: Transrail at 14.9x, KEC at 16.4x, Kalpataru at 21.4x and Techno Electric at 25.8x. That is despite the highest ROCE in the group (47.3% against 15–34%). The negative cash flow is real, but much of it is how this kind of contract works. Retention money of 10–30% per contract is held back until commissioning or the end of the defect-liability period, and it went from ₹122 crore to ₹279 crore as the business grew. Our DCF assumes 30% of every rupee of new revenue stays tied up in working capital, and still values the shares at ₹904.
• The order book supports growth even if it comes in below guidance. The ₹3,742 crore June 2026 book is 2.3x FY26 revenue, on contracts management says run 18–24 months. FY26 order inflow was ₹2,743 crore, Q1 FY27 inflow was ₹865 crore, and ₹2,200 crore of bids were awaiting results in April. Our base case assumes 30% growth in FY27, not the guided 40%. Q1 revenue of ₹437 crore is consistent with that, given that 61–71% of revenue came in the second half in each of the last two years.
• The risks the market is pricing are real, and the price allows for them. Two customers made up 56% of FY26 revenue, and 93% of the order book is in Gujarat. The promoters have a web of related-party dealings, the largest involving HKRP, which they partly own personally. The AGM sought approval for up to ₹500 crore of FY27 purchases from HKRP. Our bear case (₹541, −26%) shows what happens if these risks materialise. Our base case requires only that cash conversion improves from FY26's level, not that it becomes good.

The company reports a single segment and does not split revenue by product, state or customer type. The figures above are the closest available proxies. Note that order-book splits and revenue splits are not the same thing.

How the money is made. Rajesh Power wins state utility tenders, mainly from Gujarat's distribution companies (UGVCL, PGVCL and others) and the state transmission utility (GETCO). It then executes them as turnkey EPC contracts: it buys cables, ring main units, transformers and switchgear, and lays, installs and commissions them with its own crews. Materials made up 82% of FY26 revenue; the gross value added after materials was 18.3%. Two features differ from overhead transmission EPC. Underground work runs on existing road "right of use", which avoids most land-acquisition delays. And projects typically finish in 18–24 months. The trade-off is working capital. Utilities hold back 10–30% of each bill as retention, releasing part on installation and the rest on commissioning against a bank guarantee. Bank guarantees outstanding were ₹312.7 crore at March 2026, 80% of net worth.

Moat: real, but local.

• A specialist niche with few qualified bidders. Management counts 4–6 competitors per distribution tender and 3–4 per transmission tender. It says the company has completed or is executing more than 51,000 km of MVCC lines, which it describes as among the largest deployments in India. The figure is not independently verified.
• Deep relationships with Gujarat utilities. These are a genuine advantage, but they are also the concentration risk. Outside Gujarat the company has a handful of Rajasthan projects and its first Odisha order (₹212 crore, OPTCL, Q1 FY27).
• No pricing power. Work is won through competitive tenders. Management says escalation clauses pass "most items" through back-to-back. FY26 margins held through copper and aluminium price increases, which supports that, but FY26 also included ₹21.6 crore of discounts and rebates, against nil in FY25.
• Technology (HKRP). HKRP provides SCADA and IoT software and has put more than 1,500 GETCO distribution substations on one platform. RPSL accounts for its 25.48% stake as an associate (₹5.7 crore share of profit in FY26). HKRP is also a major supplier to RPSL, which raises the governance questions covered below.

*WACC of ~14% is a Dart Consultants assumption for an A- rated BSE SME company, not a sourced figure. CRISIL figures are from its 14 July 2026 rationale.

Margins. EBITDA margin has been steady at 11–12% for three years, and CRISIL expects it to stay there. The H2 FY26 dip to 11.4% was put down to billing mix. Operating leverage is real: employee costs fell from 4.1% to 3.6% of revenue. But other expenses rose 47%, including bank guarantee commission up 3x to ₹4.0 crore, legal and professional fees, and site costs. Nothing in the disclosures suggests margins can expand much from here, and management only commits to "a stabilized margin".

Cash and returns. The high ROCE and the negative cash flow describe the same business. Capital employed is small relative to revenue because suppliers and promoters fund much of the working capital, which makes ROCE look high. But each step up in revenue ties up more receivables and retention. Management told analysts the March 2026 receivables spike reflected "very high billing in the month of March". It said payment terms are 45–60 days and that ₹150 crore of the ₹348.8 crore had been collected by 23 April. If correct, the H1 FY27 balance sheet should show it.

Near-term catalysts (6–12 months)

• H1 FY27 results (half-year to September 2026). This is the first test of whether receivable days fall back toward the ~60 days management indicated, and whether operating cash flow turns positive.
• Results on the ₹2,200 crore of bids pending in April. Management expected outcomes by late May or June. Q1 FY27 inflow of ₹865 crore suggests some of them have converted. The FY27 targets are ₹4,000–5,000 crore of inflow and a closing order book above ₹5,000 crore.
• Commissioning of the first 400 kV GIS substation (expected September–October 2026). This would open up higher-voltage tenders.
• A CRISIL upgrade. CRISIL's stated trigger is revenue above ₹2,000 crore with a geographically diversified order book and a shorter working-capital cycle.

Longer-term catalysts (2–3 years)

• BESS commissioning (target September 2027). It would bring about ₹14.7 crore a year of contracted revenue for 12 years, at a 10–12% targeted IRR, financed 70–80% with debt. More important strategically, it is meant to qualify the company for BESS EPC tenders, where management says the substation is about 30% of project value.
• Expansion of transmission work outside Gujarat (Odisha, Rajasthan, Maharashtra, Jharkhand), and adoption of MVCC by other states.
• A larger opportunity in Gujarat. Management cites GETCO capex of about ₹9,000–10,000 crore a year, of which it sees 40–50% as addressable, plus about ₹5,000 crore a year on the distribution side.

• Three straight years of negative operating cash flow. Cumulative standalone CFO for FY24–FY26 was about −₹90 crore, against cumulative profit of about ₹251 crore. Growth explains part of this; retention and receivables explain the rest. Until a half-year shows positive cash flow, the profit figures carry less weight than they appear to.
• Payables have been stretched hard, including payables to small suppliers. MSME dues rose from ₹4.9 crore to ₹86.8 crore. Management said the suppliers concerned had "specific agreement" to terms longer than 45 days. The MSMED Act caps agreed payment terms at 45 days, and Note 44 shows no interest accrued on these dues. Late MSME payments can also be disallowed for tax purposes under Section 43B(h) of the Income-tax Act. Separately, ₹86.5 crore of payables are under bank supplier-finance arrangements with 60–120 day terms. These cut reported working capital without changing the underlying cycle.
• Heavy concentration. Two customers supplied 56% of FY26 revenue; one supplied 53% in FY25. CRISIL puts 93% of the order book in Gujarat and 90% with six customers. A change in Gujarat's capex priorities or in discom payment behaviour would hit almost the whole business.
• Order-book figures need care. The ₹3,326 crore March figure "includes L1" positions, about ₹210 crore of them according to management. Note 28 of the accounts nonetheless discloses the same ₹3,326 crore as unsatisfied contractual performance obligations. CRISIL's June figure for unexecuted orders is ₹3,164 crore, while the company reported ₹3,742 crore including L1. The FY26 target of a ₹4,500 crore order book was missed; management put this down to tender timing.
• The annual report contradicts itself repeatedly.
  • Director resignations. The Board's Report says no director resigned in FY26, but Annexure E says "four Executive Directors resigned" during the year; Note 46 dates those resignations to July 2024.
  • Associate status. The CARO report says the company has no associate companies, while HKRP is accounted for as one. Note 6 and Form AOC-1 call it a joint venture.
  • Material related-party transactions. Form AOC-2 is blank, although the Board's Report says the material transactions are reported there.
  • The ₹150 crore Rajesh Power Projects approval. The AGM notice calls it a "sale" in one table and a "purchase" in the annexure.
  • Headcount. The report gives "1,500+" employees in one place and 2,134 permanent employees in another.
  • Return on equity. It appears as 35.26%, 42.68% and 42.59% on different pages.
  None of these changes the numbers, but taken together they suggest the document was not carefully checked.

Multiples against peers. Prices are 9 October 2026 closes from screener.in. None of the peers is a clean match: they are larger, mainboard-listed and more diversified, and only Transrail is close in business mix. There is no broker coverage of Rajesh Power. Its trading history is under two years and covers a 50% fall in the share price, so we do not use a historical multiple.

Method 1: forward P/E scenarios (FY27E). We start from FY26 revenue of ₹1,628 crore, add about ₹6 crore of HKRP associate profit, take interest at about ₹15 crore (higher borrowing) and other income at ₹5 crore, and apply a 26% tax rate. EPS uses 1.80 crore shares. These are our assumptions. We set the target multiple well below every peer to reflect the cash conversion, concentration and governance issues described above.

The bear case assumes growth close to Q1 FY27's annualised run-rate (₹437cr × 4 ≈ ₹1,750cr) plus a little more, with margin slipping below the guided range, and a multiple that treats earnings as low-quality. The base case is a third below guided growth, at the middle of the margin range. A 10x multiple is a third below Transrail and less than half of Kalpataru and Techno. The bull case takes guidance at face value and gives a multiple still below every peer.

Method 2: discounted cash flow, with working capital made explicit. We use a five-year forecast (FY27–FY31) with revenue growth of 30%, 25%, 20%, 15% and 12%, an 11.5% EBITDA margin, a 26% tax rate and ₹8 crore of annual capex. We deduct ₹30 crore of BESS equity in FY27. Working capital absorbs 30% of each year's increase in revenue: worse than CRISIL's 83-day net cycle implies, but better than FY26's 38%. Discounting at a 14% WACC with 6% terminal growth and deducting ₹126 crore of net debt gives ₹904 per share (+23.3%). Free cash flow is slightly negative in FY27 (−₹4 crore) and turns positive from FY28. The terminal value is 82% of enterprise value.

Working capital is the variable that matters most. At FY26's actual intensity the DCF falls 5% below today's price. The bear and bull DCF cases give ₹554 and ₹1,410.

Target. The average of the two base cases (₹986 and ₹904) is ₹945, 28.9% above the current price. That is a BUY under Dart's rating bands (+15% or more). We call it a high-risk BUY because the two scenarios that would make it wrong are plausible. One is working capital staying at FY26 intensity; the other is the related-party limits being used at scale. Both would show up in the H1 FY27 accounts.

• 1. Working capital does not ease (downside, and the main one). If retention and receivables keep rising faster than revenue, earnings growth will need more debt and supplier credit. Our DCF falls to ₹695 at FY26's working-capital intensity. Growth funded this way would also be fragile if Gujarat discoms' payments slowed.
• 2. Value leaks to related parties (downside). If the ₹500 crore HKRP limit, the Section 185 lending powers or the sister-concern guarantees are used heavily, part of the earnings we are valuing would accrue outside the listed company, or would put its balance sheet at risk. This is not in our numbers. It is the main reason for our 10x multiple.
• 3. Concentration (downside), or diversification (upside). A slowdown in Gujarat utility capex would hit 90%+ of the order book. Conversely, if the company wins the ₹2,200 crore of pending bids and builds a real non-Gujarat transmission book, the bull case (₹1,443) and a rating upgrade become realistic.

Sources: Rajesh Power Services Annual Report 2025-26 (user-supplied, 269 pages: corporate overview, MD&A, AGM notice and related-party annexures, Board's Report and annexures, standalone and consolidated financial statements with notes 1–57, auditor's reports and CARO); H2 FY26 earnings call transcript (23 April 2026, rajeshpower.com); CRISIL Ratings rationale (14 July 2026); Q1 FY27 business update coverage (businessupturn.com, sahi.com); screener.in (prices, history and peers, 9 October 2026 close). All accessed 11 October 2026.